marf.FeatureExtraction.LPC
Class LPC

java.lang.Object
  |
  +--marf.FeatureExtraction.FeatureExtraction
        |
        +--marf.FeatureExtraction.LPC.LPC
All Implemented Interfaces:
StorageManager

public class LPC
extends FeatureExtraction

Class LPC

Implements Linear Predictive Coding

$Header: /cvsroot/marf/marf/src/marf/FeatureExtraction/LPC/LPC.java,v 1.26.2.1 2003/02/16 18:26:46 mokhov Exp $

Author:
Ian Clement

Fields inherited from class marf.FeatureExtraction.FeatureExtraction
adFeatures, oPreprocessing
 
Constructor Summary
LPC(Preprocessing poPreprocessing)
          LPC Constructor
 
Method Summary
static double autocorrelation(double[] input, int x)
          Implements the least-square autocorrelation method
static void DoLPC(double[] input, double[] output, double[] error, int p)
          Does LPC algorithm NOTE: input is assumed to be windowed, ie: input.length = N
 void dump()
          Not Implemented
 boolean extractFeatures()
          LPC Implementation of extractFeatures()
 void restore()
          Not Implemented
 
Methods inherited from class marf.FeatureExtraction.FeatureExtraction
getFeaturesArray, hamming, hamming
 
Methods inherited from class java.lang.Object
, clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

LPC

public LPC(Preprocessing poPreprocessing)
LPC Constructor
Parameters:
poPreprocessing - Preprocessing module reference
Method Detail

extractFeatures

public final boolean extractFeatures()
                              throws FeatureExtractionException
LPC Implementation of extractFeatures()
Overrides:
extractFeatures in class FeatureExtraction
Returns:
true if features were extracted, false otherwise

DoLPC

public static final void DoLPC(double[] input,
                               double[] output,
                               double[] error,
                               int p)
                        throws FeatureExtractionException
Does LPC algorithm NOTE: input is assumed to be windowed, ie: input.length = N
Parameters:
input - windowed part of incoming sample
output - resulting LPC coefficiencies
error - output LPC error
p - number of poles

autocorrelation

public static final double autocorrelation(double[] input,
                                           int x)
Implements the least-square autocorrelation method
Parameters:
input - windowed input signal
x - coefficient number
Returns:
double - correlation number

dump

public void dump()
          throws java.io.IOException
Not Implemented
Following copied from interface: marf.Storage.StorageManager
Throws:
java.io.IOException -  

restore

public void restore()
             throws java.io.IOException
Not Implemented
Following copied from interface: marf.Storage.StorageManager
Throws:
java.io.IOException -